Mortality from esophageal adenocarcinoma (EA) has risen dramatically over four decades. Efforts towards prevention and control have reflected the prevailing paradigm that Barrett's esophagus (BE) arises as a complication of reflux and greatly predisposes to EA; consequently medical care has focused on identifying persons with BE from among those with reflux, enrolling them in long-term surveillance, and aggressively treating them before they develop invasive cancer. However, the continued increase in EA mortality, coupled with the fact that 95% of BE patients will never progress to cancer, underscores the ineffectiveness of this approach. Project 2 builds directly on recent advances made by the P01 team to address critical barriers to a more effective program of prevention and control. In the past 5 years, we have published or submitted 41 research manuscripts addressing Project 2 issues and aims. During the renewal period we will utilize the rich and comprehensive data resources ofthe POl (including host and environmental risk factors, medical records, serum assays, high-density genotyping, and esophageal biopsies) collected prospectively on a BE cohort of 600 participants with 77 EA outcomes. In aim 1 we will develop risk and prediction models tailored to the absolute risk of EA that will be useful for clinical decision-making. In aim 2 we will evaluate summary genome-wide measures of somatic genomic alterations (SGA) with regard to their added predictive value. In an exploratory aim we will examine the effects of a controversial new treatment, radio frequency ablation, on genomic abnormalities in high-risk patients. In aim 3 we will consolidate the results from aim 1 with the SGA platform from aim 2, and use computer simulation to take into account time-dependent disease dynamics to develop and evaluate an optimized surveillance and prevention protocol. Results will be shared with investigators involved with risk assessment of eariier stages using resources of the BEACON consortium, and improved over time by cross-validation with other cohorts as they develop. Project 2 seeks to consolidate and extend key discoveries over the entire period of the POl to advance clinical practice and improve EA prevention and control. RELEVANCE (See instructions): The incidence of esophageal adenocarcinoma is rising faster than any other cancer. We propose to address important barriers to prevention and control of this cancer by developing and making broadly available a set of clinical tools for risk prediction and surveillance of Barrett's esophagus to be validated and applied in public health, clinical and research settings.